netdev - Re: broken ipv6 tcp csum offload on thunderx

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <CA+sq2Cdiy2HCOTVRTxJgz+qOd9Zf=px4qQmg7PmOi3LA4je4pw@mail.gmail.com>
Date:   Wed, 22 Nov 2017 14:36:08 +0530
From:   Sunil Kovvuri <sunil.kovvuri@...il.com>
To:     Florian Westphal <fw@...len.de>
Cc:     Thanneeru Srinivasulu <tsrinivasulu@...ium.com>,
        Linux Netdev List <netdev@...r.kernel.org>,
        Sunil Goutham <sgoutham@...ium.com>
Subject: Re: broken ipv6 tcp csum offload on thunderx

On Wed, Nov 22, 2017 at 2:24 PM, Florian Westphal <fw@...len.de> wrote:
> Hi.
>
> We are experiencing broken ipv6 connectivity with 4.14 kernel
> on arm64 with thunderx.
>
> ping6 still works, but it looks like tcp syn packets get sent
> with a wrong checksum -- socket remains in SYN-SENT state.
>
> after running
>
> ethtool -K enP2p1s0f1 tx-checksum-ipv6 off
>
> ipv6 tcp appears to works fine.
>
> # ethtool -i enP2p1s0f1
> driver: thunder-nicvf
> version: 1.0
> firmware-version:
> expansion-rom-version:
> bus-info: 0002:01:00.1
> supports-statistics: yes
> supports-test: no
> supports-eeprom-access: no
> supports-register-dump: yes
> supports-priv-flags: no
>
> [0.000000] Boot CPU: AArch64 Processor [431f0a10]
> [0.000000] Machine model: cavium,thunder-88xx
> [0.000000] efi: Getting EFI parameters from FDT:
> [0.000000] efi: EFI v2.40 by Cavium Thunder cn88xx EFI ThunderX-Firmware-Release-1.22.17 Sep 21 2017 14:26:28
> [0.000000] efi:  ACPI=0xffff0000  ACPI 2.0=0xffff0014 SMBIOS=0xffef0000  SMBIOS 3.0=0x10ffaf30000  ESRT=0x10fff673e18
>
> What other information do you need to debug this?

We have a fix ready for this, will submit upstream asap.

Thanks,
Sunil.

>
> Thanks,
>         Florian